This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
format and other warnings
[perl5.git] / pod / perlhack.pod
CommitLineData
e8cd7eae
GS
1=head1 NAME
2
3perlhack - How to hack at the Perl internals
4
5=head1 DESCRIPTION
6
7This document attempts to explain how Perl development takes place,
8and ends with some suggestions for people wanting to become bona fide
9porters.
10
11The perl5-porters mailing list is where the Perl standard distribution
12is maintained and developed. The list can get anywhere from 10 to 150
13messages a day, depending on the heatedness of the debate. Most days
14there are two or three patches, extensions, features, or bugs being
15discussed at a time.
16
f8e3975a 17A searchable archive of the list is at either:
e8cd7eae
GS
18
19 http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/
20
f8e3975a
IP
21or
22
23 http://archive.develooper.com/perl5-porters@perl.org/
24
e8cd7eae
GS
25List subscribers (the porters themselves) come in several flavours.
26Some are quiet curious lurkers, who rarely pitch in and instead watch
27the ongoing development to ensure they're forewarned of new changes or
28features in Perl. Some are representatives of vendors, who are there
29to make sure that Perl continues to compile and work on their
30platforms. Some patch any reported bug that they know how to fix,
31some are actively patching their pet area (threads, Win32, the regexp
32engine), while others seem to do nothing but complain. In other
33words, it's your usual mix of technical people.
34
35Over this group of porters presides Larry Wall. He has the final word
f6c51b38
GS
36in what does and does not change in the Perl language. Various
37releases of Perl are shepherded by a ``pumpking'', a porter
38responsible for gathering patches, deciding on a patch-by-patch
39feature-by-feature basis what will and will not go into the release.
caf100c0
JL
40For instance, Gurusamy Sarathy was the pumpking for the 5.6 release of
41Perl, and Jarkko Hietaniemi is the pumpking for the 5.8 release, and
42Hugo van der Sanden will be the pumpking for the 5.10 release.
e8cd7eae
GS
43
44In addition, various people are pumpkings for different things. For
45instance, Andy Dougherty and Jarkko Hietaniemi share the I<Configure>
caf100c0 46pumpkin.
e8cd7eae
GS
47
48Larry sees Perl development along the lines of the US government:
49there's the Legislature (the porters), the Executive branch (the
50pumpkings), and the Supreme Court (Larry). The legislature can
51discuss and submit patches to the executive branch all they like, but
52the executive branch is free to veto them. Rarely, the Supreme Court
53will side with the executive branch over the legislature, or the
54legislature over the executive branch. Mostly, however, the
55legislature and the executive branch are supposed to get along and
56work out their differences without impeachment or court cases.
57
58You might sometimes see reference to Rule 1 and Rule 2. Larry's power
59as Supreme Court is expressed in The Rules:
60
61=over 4
62
63=item 1
64
65Larry is always by definition right about how Perl should behave.
66This means he has final veto power on the core functionality.
67
68=item 2
69
70Larry is allowed to change his mind about any matter at a later date,
71regardless of whether he previously invoked Rule 1.
72
73=back
74
75Got that? Larry is always right, even when he was wrong. It's rare
76to see either Rule exercised, but they are often alluded to.
77
78New features and extensions to the language are contentious, because
79the criteria used by the pumpkings, Larry, and other porters to decide
80which features should be implemented and incorporated are not codified
81in a few small design goals as with some other languages. Instead,
82the heuristics are flexible and often difficult to fathom. Here is
83one person's list, roughly in decreasing order of importance, of
84heuristics that new features have to be weighed against:
85
86=over 4
87
88=item Does concept match the general goals of Perl?
89
90These haven't been written anywhere in stone, but one approximation
91is:
92
93 1. Keep it fast, simple, and useful.
94 2. Keep features/concepts as orthogonal as possible.
95 3. No arbitrary limits (platforms, data sizes, cultures).
96 4. Keep it open and exciting to use/patch/advocate Perl everywhere.
97 5. Either assimilate new technologies, or build bridges to them.
98
99=item Where is the implementation?
100
101All the talk in the world is useless without an implementation. In
102almost every case, the person or people who argue for a new feature
103will be expected to be the ones who implement it. Porters capable
104of coding new features have their own agendas, and are not available
105to implement your (possibly good) idea.
106
107=item Backwards compatibility
108
109It's a cardinal sin to break existing Perl programs. New warnings are
110contentious--some say that a program that emits warnings is not
111broken, while others say it is. Adding keywords has the potential to
112break programs, changing the meaning of existing token sequences or
113functions might break programs.
114
115=item Could it be a module instead?
116
117Perl 5 has extension mechanisms, modules and XS, specifically to avoid
118the need to keep changing the Perl interpreter. You can write modules
119that export functions, you can give those functions prototypes so they
120can be called like built-in functions, you can even write XS code to
121mess with the runtime data structures of the Perl interpreter if you
122want to implement really complicated things. If it can be done in a
123module instead of in the core, it's highly unlikely to be added.
124
125=item Is the feature generic enough?
126
127Is this something that only the submitter wants added to the language,
128or would it be broadly useful? Sometimes, instead of adding a feature
129with a tight focus, the porters might decide to wait until someone
130implements the more generalized feature. For instance, instead of
131implementing a ``delayed evaluation'' feature, the porters are waiting
132for a macro system that would permit delayed evaluation and much more.
133
134=item Does it potentially introduce new bugs?
135
136Radical rewrites of large chunks of the Perl interpreter have the
137potential to introduce new bugs. The smaller and more localized the
138change, the better.
139
140=item Does it preclude other desirable features?
141
142A patch is likely to be rejected if it closes off future avenues of
143development. For instance, a patch that placed a true and final
144interpretation on prototypes is likely to be rejected because there
145are still options for the future of prototypes that haven't been
146addressed.
147
148=item Is the implementation robust?
149
150Good patches (tight code, complete, correct) stand more chance of
151going in. Sloppy or incorrect patches might be placed on the back
152burner until the pumpking has time to fix, or might be discarded
153altogether without further notice.
154
155=item Is the implementation generic enough to be portable?
156
157The worst patches make use of a system-specific features. It's highly
158unlikely that nonportable additions to the Perl language will be
159accepted.
160
a936dd3c
NC
161=item Is the implementation tested?
162
163Patches which change behaviour (fixing bugs or introducing new features)
164must include regression tests to verify that everything works as expected.
165Without tests provided by the original author, how can anyone else changing
166perl in the future be sure that they haven't unwittingly broken the behaviour
167the patch implements? And without tests, how can the patch's author be
9d077eaa 168confident that his/her hard work put into the patch won't be accidentally
a936dd3c
NC
169thrown away by someone in the future?
170
e8cd7eae
GS
171=item Is there enough documentation?
172
173Patches without documentation are probably ill-thought out or
174incomplete. Nothing can be added without documentation, so submitting
175a patch for the appropriate manpages as well as the source code is
a936dd3c 176always a good idea.
e8cd7eae
GS
177
178=item Is there another way to do it?
179
180Larry said ``Although the Perl Slogan is I<There's More Than One Way
181to Do It>, I hesitate to make 10 ways to do something''. This is a
182tricky heuristic to navigate, though--one man's essential addition is
183another man's pointless cruft.
184
185=item Does it create too much work?
186
187Work for the pumpking, work for Perl programmers, work for module
188authors, ... Perl is supposed to be easy.
189
f6c51b38
GS
190=item Patches speak louder than words
191
192Working code is always preferred to pie-in-the-sky ideas. A patch to
193add a feature stands a much higher chance of making it to the language
194than does a random feature request, no matter how fervently argued the
195request might be. This ties into ``Will it be useful?'', as the fact
196that someone took the time to make the patch demonstrates a strong
197desire for the feature.
198
e8cd7eae
GS
199=back
200
201If you're on the list, you might hear the word ``core'' bandied
202around. It refers to the standard distribution. ``Hacking on the
203core'' means you're changing the C source code to the Perl
204interpreter. ``A core module'' is one that ships with Perl.
205
a1f349fd
MB
206=head2 Keeping in sync
207
e8cd7eae 208The source code to the Perl interpreter, in its different versions, is
f224927c
JH
209kept in a repository managed by a revision control system ( which is
210currently the Perforce program, see http://perforce.com/ ). The
e8cd7eae
GS
211pumpkings and a few others have access to the repository to check in
212changes. Periodically the pumpking for the development version of Perl
213will release a new version, so the rest of the porters can see what's
2be4c08b
GS
214changed. The current state of the main trunk of repository, and patches
215that describe the individual changes that have happened since the last
216public release are available at this location:
217
218 ftp://ftp.linux.activestate.com/pub/staff/gsar/APC/
219
a1f349fd
MB
220If you are a member of the perl5-porters mailing list, it is a good
221thing to keep in touch with the most recent changes. If not only to
222verify if what you would have posted as a bug report isn't already
223solved in the most recent available perl development branch, also
224known as perl-current, bleading edge perl, bleedperl or bleadperl.
2be4c08b
GS
225
226Needless to say, the source code in perl-current is usually in a perpetual
227state of evolution. You should expect it to be very buggy. Do B<not> use
228it for any purpose other than testing and development.
e8cd7eae 229
3e148164
JH
230Keeping in sync with the most recent branch can be done in several ways,
231but the most convenient and reliable way is using B<rsync>, available at
232ftp://rsync.samba.org/pub/rsync/ . (You can also get the most recent
233branch by FTP.)
a1f349fd
MB
234
235If you choose to keep in sync using rsync, there are two approaches
3e148164 236to doing so:
a1f349fd
MB
237
238=over 4
239
240=item rsync'ing the source tree
241
3e148164 242Presuming you are in the directory where your perl source resides
a1f349fd
MB
243and you have rsync installed and available, you can `upgrade' to
244the bleadperl using:
245
246 # rsync -avz rsync://ftp.linux.activestate.com/perl-current/ .
247
248This takes care of updating every single item in the source tree to
249the latest applied patch level, creating files that are new (to your
250distribution) and setting date/time stamps of existing files to
251reflect the bleadperl status.
252
c6d0653e
LC
253Note that this will not delete any files that were in '.' before
254the rsync. Once you are sure that the rsync is running correctly,
255run it with the --delete and the --dry-run options like this:
256
257 # rsync -avz --delete --dry-run rsync://ftp.linux.activestate.com/perl-current/ .
258
259This will I<simulate> an rsync run that also deletes files not
260present in the bleadperl master copy. Observe the results from
261this run closely. If you are sure that the actual run would delete
262no files precious to you, you could remove the '--dry-run' option.
263
a1f349fd
MB
264You can than check what patch was the latest that was applied by
265looking in the file B<.patch>, which will show the number of the
266latest patch.
267
268If you have more than one machine to keep in sync, and not all of
269them have access to the WAN (so you are not able to rsync all the
270source trees to the real source), there are some ways to get around
271this problem.
272
273=over 4
274
275=item Using rsync over the LAN
276
277Set up a local rsync server which makes the rsynced source tree
3e148164 278available to the LAN and sync the other machines against this
a1f349fd
MB
279directory.
280
1577cd80 281From http://rsync.samba.org/README.html :
a1f349fd
MB
282
283 "Rsync uses rsh or ssh for communication. It does not need to be
284 setuid and requires no special privileges for installation. It
3958b146 285 does not require an inetd entry or a daemon. You must, however,
a1f349fd
MB
286 have a working rsh or ssh system. Using ssh is recommended for
287 its security features."
288
289=item Using pushing over the NFS
290
291Having the other systems mounted over the NFS, you can take an
3e148164
JH
292active pushing approach by checking the just updated tree against
293the other not-yet synced trees. An example would be
294
295 #!/usr/bin/perl -w
296
297 use strict;
298 use File::Copy;
299
300 my %MF = map {
301 m/(\S+)/;
302 $1 => [ (stat $1)[2, 7, 9] ]; # mode, size, mtime
303 } `cat MANIFEST`;
304
305 my %remote = map { $_ => "/$_/pro/3gl/CPAN/perl-5.7.1" } qw(host1 host2);
306
307 foreach my $host (keys %remote) {
308 unless (-d $remote{$host}) {
309 print STDERR "Cannot Xsync for host $host\n";
310 next;
311 }
312 foreach my $file (keys %MF) {
313 my $rfile = "$remote{$host}/$file";
314 my ($mode, $size, $mtime) = (stat $rfile)[2, 7, 9];
315 defined $size or ($mode, $size, $mtime) = (0, 0, 0);
316 $size == $MF{$file}[1] && $mtime == $MF{$file}[2] and next;
317 printf "%4s %-34s %8d %9d %8d %9d\n",
318 $host, $file, $MF{$file}[1], $MF{$file}[2], $size, $mtime;
319 unlink $rfile;
320 copy ($file, $rfile);
321 utime time, $MF{$file}[2], $rfile;
322 chmod $MF{$file}[0], $rfile;
323 }
324 }
325
326though this is not perfect. It could be improved with checking
a1f349fd
MB
327file checksums before updating. Not all NFS systems support
328reliable utime support (when used over the NFS).
329
330=back
331
332=item rsync'ing the patches
333
334The source tree is maintained by the pumpking who applies patches to
335the files in the tree. These patches are either created by the
336pumpking himself using C<diff -c> after updating the file manually or
337by applying patches sent in by posters on the perl5-porters list.
338These patches are also saved and rsync'able, so you can apply them
339yourself to the source files.
340
341Presuming you are in a directory where your patches reside, you can
3e148164 342get them in sync with
a1f349fd
MB
343
344 # rsync -avz rsync://ftp.linux.activestate.com/perl-current-diffs/ .
345
346This makes sure the latest available patch is downloaded to your
347patch directory.
348
3e148164 349It's then up to you to apply these patches, using something like
a1f349fd 350
df3477ff 351 # last=`ls -t *.gz | sed q`
a1f349fd
MB
352 # rsync -avz rsync://ftp.linux.activestate.com/perl-current-diffs/ .
353 # find . -name '*.gz' -newer $last -exec gzcat {} \; >blead.patch
354 # cd ../perl-current
355 # patch -p1 -N <../perl-current-diffs/blead.patch
356
357or, since this is only a hint towards how it works, use CPAN-patchaperl
358from Andreas K├Ânig to have better control over the patching process.
359
360=back
361
f7e1e956 362=head2 Why rsync the source tree
a1f349fd
MB
363
364=over 4
365
10f58044 366=item It's easier to rsync the source tree
a1f349fd
MB
367
368Since you don't have to apply the patches yourself, you are sure all
369files in the source tree are in the right state.
370
371=item It's more recent
372
373According to Gurusamy Sarathy:
374
375 "... The rsync mirror is automatic and syncs with the repository
376 every five minutes.
377
3e148164 378 "Updating the patch area still requires manual intervention
a1f349fd
MB
379 (with all the goofiness that implies, which you've noted) and
380 is typically on a daily cycle. Making this process automatic
381 is on my tuit list, but don't ask me when."
382
383=item It's more reliable
384
3e148164 385Well, since the patches are updated by hand, I don't have to say any
a1f349fd
MB
386more ... (see Sarathy's remark).
387
388=back
389
f7e1e956 390=head2 Why rsync the patches
a1f349fd
MB
391
392=over 4
393
10f58044 394=item It's easier to rsync the patches
a1f349fd
MB
395
396If you have more than one machine that you want to keep in track with
3e148164 397bleadperl, it's easier to rsync the patches only once and then apply
a1f349fd
MB
398them to all the source trees on the different machines.
399
400In case you try to keep in pace on 5 different machines, for which
401only one of them has access to the WAN, rsync'ing all the source
3e148164 402trees should than be done 5 times over the NFS. Having
a1f349fd 403rsync'ed the patches only once, I can apply them to all the source
3e148164 404trees automatically. Need you say more ;-)
a1f349fd
MB
405
406=item It's a good reference
407
408If you do not only like to have the most recent development branch,
409but also like to B<fix> bugs, or extend features, you want to dive
410into the sources. If you are a seasoned perl core diver, you don't
411need no manuals, tips, roadmaps, perlguts.pod or other aids to find
412your way around. But if you are a starter, the patches may help you
413in finding where you should start and how to change the bits that
414bug you.
415
416The file B<Changes> is updated on occasions the pumpking sees as his
417own little sync points. On those occasions, he releases a tar-ball of
418the current source tree (i.e. perl@7582.tar.gz), which will be an
419excellent point to start with when choosing to use the 'rsync the
420patches' scheme. Starting with perl@7582, which means a set of source
421files on which the latest applied patch is number 7582, you apply all
f18956b7 422succeeding patches available from then on (7583, 7584, ...).
a1f349fd
MB
423
424You can use the patches later as a kind of search archive.
425
426=over 4
427
428=item Finding a start point
429
430If you want to fix/change the behaviour of function/feature Foo, just
431scan the patches for patches that mention Foo either in the subject,
3e148164 432the comments, or the body of the fix. A good chance the patch shows
a1f349fd
MB
433you the files that are affected by that patch which are very likely
434to be the starting point of your journey into the guts of perl.
435
436=item Finding how to fix a bug
437
438If you've found I<where> the function/feature Foo misbehaves, but you
439don't know how to fix it (but you do know the change you want to
440make), you can, again, peruse the patches for similar changes and
441look how others apply the fix.
442
443=item Finding the source of misbehaviour
444
445When you keep in sync with bleadperl, the pumpking would love to
3958b146 446I<see> that the community efforts really work. So after each of his
a1f349fd
MB
447sync points, you are to 'make test' to check if everything is still
448in working order. If it is, you do 'make ok', which will send an OK
449report to perlbug@perl.org. (If you do not have access to a mailer
3e148164 450from the system you just finished successfully 'make test', you can
a1f349fd
MB
451do 'make okfile', which creates the file C<perl.ok>, which you can
452than take to your favourite mailer and mail yourself).
453
3958b146 454But of course, as always, things will not always lead to a success
a1f349fd
MB
455path, and one or more test do not pass the 'make test'. Before
456sending in a bug report (using 'make nok' or 'make nokfile'), check
457the mailing list if someone else has reported the bug already and if
458so, confirm it by replying to that message. If not, you might want to
459trace the source of that misbehaviour B<before> sending in the bug,
460which will help all the other porters in finding the solution.
461
3e148164
JH
462Here the saved patches come in very handy. You can check the list of
463patches to see which patch changed what file and what change caused
464the misbehaviour. If you note that in the bug report, it saves the
465one trying to solve it, looking for that point.
a1f349fd
MB
466
467=back
468
469If searching the patches is too bothersome, you might consider using
470perl's bugtron to find more information about discussions and
471ramblings on posted bugs.
472
3e148164
JH
473If you want to get the best of both worlds, rsync both the source
474tree for convenience, reliability and ease and rsync the patches
475for reference.
476
52315700
RF
477=back
478
479
480=head2 Perlbug remote interface
481
482=over 4
483
f224927c
JH
484There are three (3) remote administrative interfaces for modifying bug
485status, category, etc. In all cases an admin must be first registered
486with the Perlbug database by sending an email request to
487richard@perl.org or bugmongers@perl.org.
52315700 488
f224927c
JH
489The main requirement is the willingness to classify, (with the
490emphasis on closing where possible :), outstanding bugs. Further
491explanation can be garnered from the web at http://bugs.perl.org/ , or
492by asking on the admin mailing list at: bugmongers@perl.org
52315700
RF
493
494For more info on the web see
495
496 http://bugs.perl.org/perlbug.cgi?req=spec
497
498
499B<The interfaces:>
500
501
502=item 1 http://bugs.perl.org
503
504Login via the web, (remove B<admin/> if only browsing), where interested Cc's, tests, patches and change-ids, etc. may be assigned.
505
506 http://bugs.perl.org/admin/index.html
507
508
509=item 2 bugdb@perl.org
510
511Where the subject line is used for commands:
512
513 To: bugdb@perl.org
514 Subject: -a close bugid1 bugid2 aix install
515
516 To: bugdb@perl.org
517 Subject: -h
518
519
520=item 3 commands_and_bugdids@bugs.perl.org
521
522Where the address itself is the source for the commands:
523
524 To: close_bugid1_bugid2_aix@bugs.perl.org
525
526 To: help@bugs.perl.org
527
528
529=item notes, patches, tests
530
531For patches and tests, the message body is assigned to the appropriate bug/s and forwarded to p5p for their attention.
532
533 To: test_<bugid1>_aix_close@bugs.perl.org
534 Subject: this is a test for the (now closed) aix bug
535
536 Test is the body of the mail
537
538=back
539
a1f349fd
MB
540=head2 Submitting patches
541
f7e1e956
MS
542Always submit patches to I<perl5-porters@perl.org>. If you're
543patching a core module and there's an author listed, send the author a
544copy (see L<Patching a core module>). This lets other porters review
545your patch, which catches a surprising number of errors in patches.
546Either use the diff program (available in source code form from
f224927c 547ftp://ftp.gnu.org/pub/gnu/ , or use Johan Vromans' I<makepatch>
f7e1e956
MS
548(available from I<CPAN/authors/id/JV/>). Unified diffs are preferred,
549but context diffs are accepted. Do not send RCS-style diffs or diffs
550without context lines. More information is given in the
551I<Porting/patching.pod> file in the Perl source distribution. Please
552patch against the latest B<development> version (e.g., if you're
553fixing a bug in the 5.005 track, patch against the latest 5.005_5x
554version). Only patches that survive the heat of the development
555branch get applied to maintenance versions.
556
557Your patch should update the documentation and test suite. See
558L<Writing a test>.
e8cd7eae
GS
559
560To report a bug in Perl, use the program I<perlbug> which comes with
561Perl (if you can't get Perl to work, send mail to the address
f18956b7 562I<perlbug@perl.org> or I<perlbug@perl.com>). Reporting bugs through
e8cd7eae 563I<perlbug> feeds into the automated bug-tracking system, access to
f224927c 564which is provided through the web at http://bugs.perl.org/ . It
e8cd7eae
GS
565often pays to check the archives of the perl5-porters mailing list to
566see whether the bug you're reporting has been reported before, and if
567so whether it was considered a bug. See above for the location of
568the searchable archives.
569
f224927c 570The CPAN testers ( http://testers.cpan.org/ ) are a group of
e8cd7eae 571volunteers who test CPAN modules on a variety of platforms. Perl Labs
f224927c 572( http://labs.perl.org/ ) automatically tests Perl source releases on
f6c51b38
GS
573platforms and gives feedback to the CPAN testers mailing list. Both
574efforts welcome volunteers.
e8cd7eae 575
e8cd7eae
GS
576It's a good idea to read and lurk for a while before chipping in.
577That way you'll get to see the dynamic of the conversations, learn the
578personalities of the players, and hopefully be better prepared to make
579a useful contribution when do you speak up.
580
581If after all this you still think you want to join the perl5-porters
f6c51b38
GS
582mailing list, send mail to I<perl5-porters-subscribe@perl.org>. To
583unsubscribe, send mail to I<perl5-porters-unsubscribe@perl.org>.
e8cd7eae 584
a422fd2d
SC
585To hack on the Perl guts, you'll need to read the following things:
586
587=over 3
588
589=item L<perlguts>
590
591This is of paramount importance, since it's the documentation of what
592goes where in the Perl source. Read it over a couple of times and it
593might start to make sense - don't worry if it doesn't yet, because the
594best way to study it is to read it in conjunction with poking at Perl
595source, and we'll do that later on.
596
597You might also want to look at Gisle Aas's illustrated perlguts -
598there's no guarantee that this will be absolutely up-to-date with the
599latest documentation in the Perl core, but the fundamentals will be
1577cd80 600right. ( http://gisle.aas.no/perl/illguts/ )
a422fd2d
SC
601
602=item L<perlxstut> and L<perlxs>
603
604A working knowledge of XSUB programming is incredibly useful for core
605hacking; XSUBs use techniques drawn from the PP code, the portion of the
606guts that actually executes a Perl program. It's a lot gentler to learn
607those techniques from simple examples and explanation than from the core
608itself.
609
610=item L<perlapi>
611
612The documentation for the Perl API explains what some of the internal
613functions do, as well as the many macros used in the source.
614
615=item F<Porting/pumpkin.pod>
616
617This is a collection of words of wisdom for a Perl porter; some of it is
618only useful to the pumpkin holder, but most of it applies to anyone
619wanting to go about Perl development.
620
621=item The perl5-porters FAQ
622
623This is posted to perl5-porters at the beginning on every month, and
f224927c 624should be available from http://perlhacker.org/p5p-faq ; alternatively,
a422fd2d
SC
625you can get the FAQ emailed to you by sending mail to
626C<perl5-porters-faq@perl.org>. It contains hints on reading
627perl5-porters, information on how perl5-porters works and how Perl
628development in general works.
629
630=back
631
632=head2 Finding Your Way Around
633
634Perl maintenance can be split into a number of areas, and certain people
635(pumpkins) will have responsibility for each area. These areas sometimes
636correspond to files or directories in the source kit. Among the areas are:
637
638=over 3
639
640=item Core modules
641
642Modules shipped as part of the Perl core live in the F<lib/> and F<ext/>
643subdirectories: F<lib/> is for the pure-Perl modules, and F<ext/>
644contains the core XS modules.
645
f7e1e956
MS
646=item Tests
647
648There are tests for nearly all the modules, built-ins and major bits
649of functionality. Test files all have a .t suffix. Module tests live
650in the F<lib/> and F<ext/> directories next to the module being
651tested. Others live in F<t/>. See L<Writing a test>
652
a422fd2d
SC
653=item Documentation
654
655Documentation maintenance includes looking after everything in the
656F<pod/> directory, (as well as contributing new documentation) and
657the documentation to the modules in core.
658
659=item Configure
660
661The configure process is the way we make Perl portable across the
662myriad of operating systems it supports. Responsibility for the
663configure, build and installation process, as well as the overall
664portability of the core code rests with the configure pumpkin - others
665help out with individual operating systems.
666
667The files involved are the operating system directories, (F<win32/>,
668F<os2/>, F<vms/> and so on) the shell scripts which generate F<config.h>
669and F<Makefile>, as well as the metaconfig files which generate
670F<Configure>. (metaconfig isn't included in the core distribution.)
671
672=item Interpreter
673
674And of course, there's the core of the Perl interpreter itself. Let's
675have a look at that in a little more detail.
676
677=back
678
679Before we leave looking at the layout, though, don't forget that
680F<MANIFEST> contains not only the file names in the Perl distribution,
681but short descriptions of what's in them, too. For an overview of the
682important files, try this:
683
684 perl -lne 'print if /^[^\/]+\.[ch]\s+/' MANIFEST
685
686=head2 Elements of the interpreter
687
688The work of the interpreter has two main stages: compiling the code
689into the internal representation, or bytecode, and then executing it.
690L<perlguts/Compiled code> explains exactly how the compilation stage
691happens.
692
693Here is a short breakdown of perl's operation:
694
695=over 3
696
697=item Startup
698
699The action begins in F<perlmain.c>. (or F<miniperlmain.c> for miniperl)
700This is very high-level code, enough to fit on a single screen, and it
701resembles the code found in L<perlembed>; most of the real action takes
702place in F<perl.c>
703
704First, F<perlmain.c> allocates some memory and constructs a Perl
705interpreter:
706
707 1 PERL_SYS_INIT3(&argc,&argv,&env);
708 2
709 3 if (!PL_do_undump) {
710 4 my_perl = perl_alloc();
711 5 if (!my_perl)
712 6 exit(1);
713 7 perl_construct(my_perl);
714 8 PL_perl_destruct_level = 0;
715 9 }
716
717Line 1 is a macro, and its definition is dependent on your operating
718system. Line 3 references C<PL_do_undump>, a global variable - all
719global variables in Perl start with C<PL_>. This tells you whether the
720current running program was created with the C<-u> flag to perl and then
721F<undump>, which means it's going to be false in any sane context.
722
723Line 4 calls a function in F<perl.c> to allocate memory for a Perl
724interpreter. It's quite a simple function, and the guts of it looks like
725this:
726
727 my_perl = (PerlInterpreter*)PerlMem_malloc(sizeof(PerlInterpreter));
728
729Here you see an example of Perl's system abstraction, which we'll see
730later: C<PerlMem_malloc> is either your system's C<malloc>, or Perl's
731own C<malloc> as defined in F<malloc.c> if you selected that option at
732configure time.
733
734Next, in line 7, we construct the interpreter; this sets up all the
735special variables that Perl needs, the stacks, and so on.
736
737Now we pass Perl the command line options, and tell it to go:
738
739 exitstatus = perl_parse(my_perl, xs_init, argc, argv, (char **)NULL);
740 if (!exitstatus) {
741 exitstatus = perl_run(my_perl);
742 }
743
744
745C<perl_parse> is actually a wrapper around C<S_parse_body>, as defined
746in F<perl.c>, which processes the command line options, sets up any
747statically linked XS modules, opens the program and calls C<yyparse> to
748parse it.
749
750=item Parsing
751
752The aim of this stage is to take the Perl source, and turn it into an op
753tree. We'll see what one of those looks like later. Strictly speaking,
754there's three things going on here.
755
756C<yyparse>, the parser, lives in F<perly.c>, although you're better off
757reading the original YACC input in F<perly.y>. (Yes, Virginia, there
758B<is> a YACC grammar for Perl!) The job of the parser is to take your
759code and `understand' it, splitting it into sentences, deciding which
760operands go with which operators and so on.
761
762The parser is nobly assisted by the lexer, which chunks up your input
763into tokens, and decides what type of thing each token is: a variable
764name, an operator, a bareword, a subroutine, a core function, and so on.
765The main point of entry to the lexer is C<yylex>, and that and its
766associated routines can be found in F<toke.c>. Perl isn't much like
767other computer languages; it's highly context sensitive at times, it can
768be tricky to work out what sort of token something is, or where a token
769ends. As such, there's a lot of interplay between the tokeniser and the
770parser, which can get pretty frightening if you're not used to it.
771
772As the parser understands a Perl program, it builds up a tree of
773operations for the interpreter to perform during execution. The routines
774which construct and link together the various operations are to be found
775in F<op.c>, and will be examined later.
776
777=item Optimization
778
779Now the parsing stage is complete, and the finished tree represents
780the operations that the Perl interpreter needs to perform to execute our
781program. Next, Perl does a dry run over the tree looking for
782optimisations: constant expressions such as C<3 + 4> will be computed
783now, and the optimizer will also see if any multiple operations can be
784replaced with a single one. For instance, to fetch the variable C<$foo>,
785instead of grabbing the glob C<*foo> and looking at the scalar
786component, the optimizer fiddles the op tree to use a function which
787directly looks up the scalar in question. The main optimizer is C<peep>
788in F<op.c>, and many ops have their own optimizing functions.
789
790=item Running
791
792Now we're finally ready to go: we have compiled Perl byte code, and all
793that's left to do is run it. The actual execution is done by the
794C<runops_standard> function in F<run.c>; more specifically, it's done by
795these three innocent looking lines:
796
797 while ((PL_op = CALL_FPTR(PL_op->op_ppaddr)(aTHX))) {
798 PERL_ASYNC_CHECK();
799 }
800
801You may be more comfortable with the Perl version of that:
802
803 PERL_ASYNC_CHECK() while $Perl::op = &{$Perl::op->{function}};
804
805Well, maybe not. Anyway, each op contains a function pointer, which
806stipulates the function which will actually carry out the operation.
807This function will return the next op in the sequence - this allows for
808things like C<if> which choose the next op dynamically at run time.
809The C<PERL_ASYNC_CHECK> makes sure that things like signals interrupt
810execution if required.
811
812The actual functions called are known as PP code, and they're spread
813between four files: F<pp_hot.c> contains the `hot' code, which is most
814often used and highly optimized, F<pp_sys.c> contains all the
815system-specific functions, F<pp_ctl.c> contains the functions which
816implement control structures (C<if>, C<while> and the like) and F<pp.c>
817contains everything else. These are, if you like, the C code for Perl's
818built-in functions and operators.
819
820=back
821
822=head2 Internal Variable Types
823
824You should by now have had a look at L<perlguts>, which tells you about
825Perl's internal variable types: SVs, HVs, AVs and the rest. If not, do
826that now.
827
828These variables are used not only to represent Perl-space variables, but
829also any constants in the code, as well as some structures completely
830internal to Perl. The symbol table, for instance, is an ordinary Perl
831hash. Your code is represented by an SV as it's read into the parser;
832any program files you call are opened via ordinary Perl filehandles, and
833so on.
834
835The core L<Devel::Peek|Devel::Peek> module lets us examine SVs from a
836Perl program. Let's see, for instance, how Perl treats the constant
837C<"hello">.
838
839 % perl -MDevel::Peek -e 'Dump("hello")'
840 1 SV = PV(0xa041450) at 0xa04ecbc
841 2 REFCNT = 1
842 3 FLAGS = (POK,READONLY,pPOK)
843 4 PV = 0xa0484e0 "hello"\0
844 5 CUR = 5
845 6 LEN = 6
846
847Reading C<Devel::Peek> output takes a bit of practise, so let's go
848through it line by line.
849
850Line 1 tells us we're looking at an SV which lives at C<0xa04ecbc> in
851memory. SVs themselves are very simple structures, but they contain a
852pointer to a more complex structure. In this case, it's a PV, a
853structure which holds a string value, at location C<0xa041450>. Line 2
854is the reference count; there are no other references to this data, so
855it's 1.
856
857Line 3 are the flags for this SV - it's OK to use it as a PV, it's a
858read-only SV (because it's a constant) and the data is a PV internally.
859Next we've got the contents of the string, starting at location
860C<0xa0484e0>.
861
862Line 5 gives us the current length of the string - note that this does
863B<not> include the null terminator. Line 6 is not the length of the
864string, but the length of the currently allocated buffer; as the string
865grows, Perl automatically extends the available storage via a routine
866called C<SvGROW>.
867
868You can get at any of these quantities from C very easily; just add
869C<Sv> to the name of the field shown in the snippet, and you've got a
870macro which will return the value: C<SvCUR(sv)> returns the current
871length of the string, C<SvREFCOUNT(sv)> returns the reference count,
872C<SvPV(sv, len)> returns the string itself with its length, and so on.
873More macros to manipulate these properties can be found in L<perlguts>.
874
875Let's take an example of manipulating a PV, from C<sv_catpvn>, in F<sv.c>
876
877 1 void
878 2 Perl_sv_catpvn(pTHX_ register SV *sv, register const char *ptr, register STRLEN len)
879 3 {
880 4 STRLEN tlen;
881 5 char *junk;
882
883 6 junk = SvPV_force(sv, tlen);
884 7 SvGROW(sv, tlen + len + 1);
885 8 if (ptr == junk)
886 9 ptr = SvPVX(sv);
887 10 Move(ptr,SvPVX(sv)+tlen,len,char);
888 11 SvCUR(sv) += len;
889 12 *SvEND(sv) = '\0';
890 13 (void)SvPOK_only_UTF8(sv); /* validate pointer */
891 14 SvTAINT(sv);
892 15 }
893
894This is a function which adds a string, C<ptr>, of length C<len> onto
895the end of the PV stored in C<sv>. The first thing we do in line 6 is
896make sure that the SV B<has> a valid PV, by calling the C<SvPV_force>
897macro to force a PV. As a side effect, C<tlen> gets set to the current
898value of the PV, and the PV itself is returned to C<junk>.
899
b1866b2d 900In line 7, we make sure that the SV will have enough room to accommodate
a422fd2d
SC
901the old string, the new string and the null terminator. If C<LEN> isn't
902big enough, C<SvGROW> will reallocate space for us.
903
904Now, if C<junk> is the same as the string we're trying to add, we can
905grab the string directly from the SV; C<SvPVX> is the address of the PV
906in the SV.
907
908Line 10 does the actual catenation: the C<Move> macro moves a chunk of
909memory around: we move the string C<ptr> to the end of the PV - that's
910the start of the PV plus its current length. We're moving C<len> bytes
911of type C<char>. After doing so, we need to tell Perl we've extended the
912string, by altering C<CUR> to reflect the new length. C<SvEND> is a
913macro which gives us the end of the string, so that needs to be a
914C<"\0">.
915
916Line 13 manipulates the flags; since we've changed the PV, any IV or NV
917values will no longer be valid: if we have C<$a=10; $a.="6";> we don't
918want to use the old IV of 10. C<SvPOK_only_utf8> is a special UTF8-aware
919version of C<SvPOK_only>, a macro which turns off the IOK and NOK flags
920and turns on POK. The final C<SvTAINT> is a macro which launders tainted
921data if taint mode is turned on.
922
923AVs and HVs are more complicated, but SVs are by far the most common
924variable type being thrown around. Having seen something of how we
925manipulate these, let's go on and look at how the op tree is
926constructed.
927
928=head2 Op Trees
929
930First, what is the op tree, anyway? The op tree is the parsed
931representation of your program, as we saw in our section on parsing, and
932it's the sequence of operations that Perl goes through to execute your
933program, as we saw in L</Running>.
934
935An op is a fundamental operation that Perl can perform: all the built-in
936functions and operators are ops, and there are a series of ops which
937deal with concepts the interpreter needs internally - entering and
938leaving a block, ending a statement, fetching a variable, and so on.
939
940The op tree is connected in two ways: you can imagine that there are two
941"routes" through it, two orders in which you can traverse the tree.
942First, parse order reflects how the parser understood the code, and
943secondly, execution order tells perl what order to perform the
944operations in.
945
946The easiest way to examine the op tree is to stop Perl after it has
947finished parsing, and get it to dump out the tree. This is exactly what
948the compiler backends L<B::Terse|B::Terse> and L<B::Debug|B::Debug> do.
949
950Let's have a look at how Perl sees C<$a = $b + $c>:
951
952 % perl -MO=Terse -e '$a=$b+$c'
953 1 LISTOP (0x8179888) leave
954 2 OP (0x81798b0) enter
955 3 COP (0x8179850) nextstate
956 4 BINOP (0x8179828) sassign
957 5 BINOP (0x8179800) add [1]
958 6 UNOP (0x81796e0) null [15]
959 7 SVOP (0x80fafe0) gvsv GV (0x80fa4cc) *b
960 8 UNOP (0x81797e0) null [15]
961 9 SVOP (0x8179700) gvsv GV (0x80efeb0) *c
962 10 UNOP (0x816b4f0) null [15]
963 11 SVOP (0x816dcf0) gvsv GV (0x80fa460) *a
964
965Let's start in the middle, at line 4. This is a BINOP, a binary
966operator, which is at location C<0x8179828>. The specific operator in
967question is C<sassign> - scalar assignment - and you can find the code
968which implements it in the function C<pp_sassign> in F<pp_hot.c>. As a
969binary operator, it has two children: the add operator, providing the
970result of C<$b+$c>, is uppermost on line 5, and the left hand side is on
971line 10.
972
973Line 10 is the null op: this does exactly nothing. What is that doing
974there? If you see the null op, it's a sign that something has been
975optimized away after parsing. As we mentioned in L</Optimization>,
976the optimization stage sometimes converts two operations into one, for
977example when fetching a scalar variable. When this happens, instead of
978rewriting the op tree and cleaning up the dangling pointers, it's easier
979just to replace the redundant operation with the null op. Originally,
980the tree would have looked like this:
981
982 10 SVOP (0x816b4f0) rv2sv [15]
983 11 SVOP (0x816dcf0) gv GV (0x80fa460) *a
984
985That is, fetch the C<a> entry from the main symbol table, and then look
986at the scalar component of it: C<gvsv> (C<pp_gvsv> into F<pp_hot.c>)
987happens to do both these things.
988
989The right hand side, starting at line 5 is similar to what we've just
990seen: we have the C<add> op (C<pp_add> also in F<pp_hot.c>) add together
991two C<gvsv>s.
992
993Now, what's this about?
994
995 1 LISTOP (0x8179888) leave
996 2 OP (0x81798b0) enter
997 3 COP (0x8179850) nextstate
998
999C<enter> and C<leave> are scoping ops, and their job is to perform any
1000housekeeping every time you enter and leave a block: lexical variables
1001are tidied up, unreferenced variables are destroyed, and so on. Every
1002program will have those first three lines: C<leave> is a list, and its
1003children are all the statements in the block. Statements are delimited
1004by C<nextstate>, so a block is a collection of C<nextstate> ops, with
1005the ops to be performed for each statement being the children of
1006C<nextstate>. C<enter> is a single op which functions as a marker.
1007
1008That's how Perl parsed the program, from top to bottom:
1009
1010 Program
1011 |
1012 Statement
1013 |
1014 =
1015 / \
1016 / \
1017 $a +
1018 / \
1019 $b $c
1020
1021However, it's impossible to B<perform> the operations in this order:
1022you have to find the values of C<$b> and C<$c> before you add them
1023together, for instance. So, the other thread that runs through the op
1024tree is the execution order: each op has a field C<op_next> which points
1025to the next op to be run, so following these pointers tells us how perl
1026executes the code. We can traverse the tree in this order using
1027the C<exec> option to C<B::Terse>:
1028
1029 % perl -MO=Terse,exec -e '$a=$b+$c'
1030 1 OP (0x8179928) enter
1031 2 COP (0x81798c8) nextstate
1032 3 SVOP (0x81796c8) gvsv GV (0x80fa4d4) *b
1033 4 SVOP (0x8179798) gvsv GV (0x80efeb0) *c
1034 5 BINOP (0x8179878) add [1]
1035 6 SVOP (0x816dd38) gvsv GV (0x80fa468) *a
1036 7 BINOP (0x81798a0) sassign
1037 8 LISTOP (0x8179900) leave
1038
1039This probably makes more sense for a human: enter a block, start a
1040statement. Get the values of C<$b> and C<$c>, and add them together.
1041Find C<$a>, and assign one to the other. Then leave.
1042
1043The way Perl builds up these op trees in the parsing process can be
1044unravelled by examining F<perly.y>, the YACC grammar. Let's take the
1045piece we need to construct the tree for C<$a = $b + $c>
1046
1047 1 term : term ASSIGNOP term
1048 2 { $$ = newASSIGNOP(OPf_STACKED, $1, $2, $3); }
1049 3 | term ADDOP term
1050 4 { $$ = newBINOP($2, 0, scalar($1), scalar($3)); }
1051
1052If you're not used to reading BNF grammars, this is how it works: You're
1053fed certain things by the tokeniser, which generally end up in upper
1054case. Here, C<ADDOP>, is provided when the tokeniser sees C<+> in your
1055code. C<ASSIGNOP> is provided when C<=> is used for assigning. These are
1056`terminal symbols', because you can't get any simpler than them.
1057
1058The grammar, lines one and three of the snippet above, tells you how to
1059build up more complex forms. These complex forms, `non-terminal symbols'
1060are generally placed in lower case. C<term> here is a non-terminal
1061symbol, representing a single expression.
1062
1063The grammar gives you the following rule: you can make the thing on the
1064left of the colon if you see all the things on the right in sequence.
1065This is called a "reduction", and the aim of parsing is to completely
1066reduce the input. There are several different ways you can perform a
1067reduction, separated by vertical bars: so, C<term> followed by C<=>
1068followed by C<term> makes a C<term>, and C<term> followed by C<+>
1069followed by C<term> can also make a C<term>.
1070
1071So, if you see two terms with an C<=> or C<+>, between them, you can
1072turn them into a single expression. When you do this, you execute the
1073code in the block on the next line: if you see C<=>, you'll do the code
1074in line 2. If you see C<+>, you'll do the code in line 4. It's this code
1075which contributes to the op tree.
1076
1077 | term ADDOP term
1078 { $$ = newBINOP($2, 0, scalar($1), scalar($3)); }
1079
1080What this does is creates a new binary op, and feeds it a number of
1081variables. The variables refer to the tokens: C<$1> is the first token in
1082the input, C<$2> the second, and so on - think regular expression
1083backreferences. C<$$> is the op returned from this reduction. So, we
1084call C<newBINOP> to create a new binary operator. The first parameter to
1085C<newBINOP>, a function in F<op.c>, is the op type. It's an addition
1086operator, so we want the type to be C<ADDOP>. We could specify this
1087directly, but it's right there as the second token in the input, so we
1088use C<$2>. The second parameter is the op's flags: 0 means `nothing
1089special'. Then the things to add: the left and right hand side of our
1090expression, in scalar context.
1091
1092=head2 Stacks
1093
1094When perl executes something like C<addop>, how does it pass on its
1095results to the next op? The answer is, through the use of stacks. Perl
1096has a number of stacks to store things it's currently working on, and
1097we'll look at the three most important ones here.
1098
1099=over 3
1100
1101=item Argument stack
1102
1103Arguments are passed to PP code and returned from PP code using the
1104argument stack, C<ST>. The typical way to handle arguments is to pop
1105them off the stack, deal with them how you wish, and then push the result
1106back onto the stack. This is how, for instance, the cosine operator
1107works:
1108
1109 NV value;
1110 value = POPn;
1111 value = Perl_cos(value);
1112 XPUSHn(value);
1113
1114We'll see a more tricky example of this when we consider Perl's macros
1115below. C<POPn> gives you the NV (floating point value) of the top SV on
1116the stack: the C<$x> in C<cos($x)>. Then we compute the cosine, and push
1117the result back as an NV. The C<X> in C<XPUSHn> means that the stack
1118should be extended if necessary - it can't be necessary here, because we
1119know there's room for one more item on the stack, since we've just
1120removed one! The C<XPUSH*> macros at least guarantee safety.
1121
1122Alternatively, you can fiddle with the stack directly: C<SP> gives you
1123the first element in your portion of the stack, and C<TOP*> gives you
1124the top SV/IV/NV/etc. on the stack. So, for instance, to do unary
1125negation of an integer:
1126
1127 SETi(-TOPi);
1128
1129Just set the integer value of the top stack entry to its negation.
1130
1131Argument stack manipulation in the core is exactly the same as it is in
1132XSUBs - see L<perlxstut>, L<perlxs> and L<perlguts> for a longer
1133description of the macros used in stack manipulation.
1134
1135=item Mark stack
1136
1137I say `your portion of the stack' above because PP code doesn't
1138necessarily get the whole stack to itself: if your function calls
1139another function, you'll only want to expose the arguments aimed for the
1140called function, and not (necessarily) let it get at your own data. The
1141way we do this is to have a `virtual' bottom-of-stack, exposed to each
1142function. The mark stack keeps bookmarks to locations in the argument
1143stack usable by each function. For instance, when dealing with a tied
1144variable, (internally, something with `P' magic) Perl has to call
1145methods for accesses to the tied variables. However, we need to separate
1146the arguments exposed to the method to the argument exposed to the
1147original function - the store or fetch or whatever it may be. Here's how
1148the tied C<push> is implemented; see C<av_push> in F<av.c>:
1149
1150 1 PUSHMARK(SP);
1151 2 EXTEND(SP,2);
1152 3 PUSHs(SvTIED_obj((SV*)av, mg));
1153 4 PUSHs(val);
1154 5 PUTBACK;
1155 6 ENTER;
1156 7 call_method("PUSH", G_SCALAR|G_DISCARD);
1157 8 LEAVE;
1158 9 POPSTACK;
13a2d996 1159
a422fd2d
SC
1160The lines which concern the mark stack are the first, fifth and last
1161lines: they save away, restore and remove the current position of the
1162argument stack.
1163
1164Let's examine the whole implementation, for practice:
1165
1166 1 PUSHMARK(SP);
1167
1168Push the current state of the stack pointer onto the mark stack. This is
1169so that when we've finished adding items to the argument stack, Perl
1170knows how many things we've added recently.
1171
1172 2 EXTEND(SP,2);
1173 3 PUSHs(SvTIED_obj((SV*)av, mg));
1174 4 PUSHs(val);
1175
1176We're going to add two more items onto the argument stack: when you have
1177a tied array, the C<PUSH> subroutine receives the object and the value
1178to be pushed, and that's exactly what we have here - the tied object,
1179retrieved with C<SvTIED_obj>, and the value, the SV C<val>.
1180
1181 5 PUTBACK;
1182
1183Next we tell Perl to make the change to the global stack pointer: C<dSP>
1184only gave us a local copy, not a reference to the global.
1185
1186 6 ENTER;
1187 7 call_method("PUSH", G_SCALAR|G_DISCARD);
1188 8 LEAVE;
1189
1190C<ENTER> and C<LEAVE> localise a block of code - they make sure that all
1191variables are tidied up, everything that has been localised gets
1192its previous value returned, and so on. Think of them as the C<{> and
1193C<}> of a Perl block.
1194
1195To actually do the magic method call, we have to call a subroutine in
1196Perl space: C<call_method> takes care of that, and it's described in
1197L<perlcall>. We call the C<PUSH> method in scalar context, and we're
1198going to discard its return value.
1199
1200 9 POPSTACK;
1201
1202Finally, we remove the value we placed on the mark stack, since we
1203don't need it any more.
1204
1205=item Save stack
1206
1207C doesn't have a concept of local scope, so perl provides one. We've
1208seen that C<ENTER> and C<LEAVE> are used as scoping braces; the save
1209stack implements the C equivalent of, for example:
1210
1211 {
1212 local $foo = 42;
1213 ...
1214 }
1215
1216See L<perlguts/Localising Changes> for how to use the save stack.
1217
1218=back
1219
1220=head2 Millions of Macros
1221
1222One thing you'll notice about the Perl source is that it's full of
1223macros. Some have called the pervasive use of macros the hardest thing
1224to understand, others find it adds to clarity. Let's take an example,
1225the code which implements the addition operator:
1226
1227 1 PP(pp_add)
1228 2 {
39644a26 1229 3 dSP; dATARGET; tryAMAGICbin(add,opASSIGN);
a422fd2d
SC
1230 4 {
1231 5 dPOPTOPnnrl_ul;
1232 6 SETn( left + right );
1233 7 RETURN;
1234 8 }
1235 9 }
1236
1237Every line here (apart from the braces, of course) contains a macro. The
1238first line sets up the function declaration as Perl expects for PP code;
1239line 3 sets up variable declarations for the argument stack and the
1240target, the return value of the operation. Finally, it tries to see if
1241the addition operation is overloaded; if so, the appropriate subroutine
1242is called.
1243
1244Line 5 is another variable declaration - all variable declarations start
1245with C<d> - which pops from the top of the argument stack two NVs (hence
1246C<nn>) and puts them into the variables C<right> and C<left>, hence the
1247C<rl>. These are the two operands to the addition operator. Next, we
1248call C<SETn> to set the NV of the return value to the result of adding
1249the two values. This done, we return - the C<RETURN> macro makes sure
1250that our return value is properly handled, and we pass the next operator
1251to run back to the main run loop.
1252
1253Most of these macros are explained in L<perlapi>, and some of the more
1254important ones are explained in L<perlxs> as well. Pay special attention
1255to L<perlguts/Background and PERL_IMPLICIT_CONTEXT> for information on
1256the C<[pad]THX_?> macros.
1257
a422fd2d
SC
1258=head2 Poking at Perl
1259
1260To really poke around with Perl, you'll probably want to build Perl for
1261debugging, like this:
1262
1263 ./Configure -d -D optimize=-g
1264 make
1265
1266C<-g> is a flag to the C compiler to have it produce debugging
1267information which will allow us to step through a running program.
1268F<Configure> will also turn on the C<DEBUGGING> compilation symbol which
1269enables all the internal debugging code in Perl. There are a whole bunch
1270of things you can debug with this: L<perlrun> lists them all, and the
1271best way to find out about them is to play about with them. The most
1272useful options are probably
1273
1274 l Context (loop) stack processing
1275 t Trace execution
1276 o Method and overloading resolution
1277 c String/numeric conversions
1278
1279Some of the functionality of the debugging code can be achieved using XS
1280modules.
13a2d996 1281
a422fd2d
SC
1282 -Dr => use re 'debug'
1283 -Dx => use O 'Debug'
1284
1285=head2 Using a source-level debugger
1286
1287If the debugging output of C<-D> doesn't help you, it's time to step
1288through perl's execution with a source-level debugger.
1289
1290=over 3
1291
1292=item *
1293
1294We'll use C<gdb> for our examples here; the principles will apply to any
1295debugger, but check the manual of the one you're using.
1296
1297=back
1298
1299To fire up the debugger, type
1300
1301 gdb ./perl
1302
1303You'll want to do that in your Perl source tree so the debugger can read
1304the source code. You should see the copyright message, followed by the
1305prompt.
1306
1307 (gdb)
1308
1309C<help> will get you into the documentation, but here are the most
1310useful commands:
1311
1312=over 3
1313
1314=item run [args]
1315
1316Run the program with the given arguments.
1317
1318=item break function_name
1319
1320=item break source.c:xxx
1321
1322Tells the debugger that we'll want to pause execution when we reach
cea6626f 1323either the named function (but see L<perlguts/Internal Functions>!) or the given
a422fd2d
SC
1324line in the named source file.
1325
1326=item step
1327
1328Steps through the program a line at a time.
1329
1330=item next
1331
1332Steps through the program a line at a time, without descending into
1333functions.
1334
1335=item continue
1336
1337Run until the next breakpoint.
1338
1339=item finish
1340
1341Run until the end of the current function, then stop again.
1342
13a2d996 1343=item 'enter'
a422fd2d
SC
1344
1345Just pressing Enter will do the most recent operation again - it's a
1346blessing when stepping through miles of source code.
1347
1348=item print
1349
1350Execute the given C code and print its results. B<WARNING>: Perl makes
1351heavy use of macros, and F<gdb> is not aware of macros. You'll have to
1352substitute them yourself. So, for instance, you can't say
1353
1354 print SvPV_nolen(sv)
1355
1356but you have to say
1357
1358 print Perl_sv_2pv_nolen(sv)
1359
1360You may find it helpful to have a "macro dictionary", which you can
1361produce by saying C<cpp -dM perl.c | sort>. Even then, F<cpp> won't
1362recursively apply the macros for you.
1363
1364=back
1365
1366=head2 Dumping Perl Data Structures
1367
1368One way to get around this macro hell is to use the dumping functions in
1369F<dump.c>; these work a little like an internal
1370L<Devel::Peek|Devel::Peek>, but they also cover OPs and other structures
1371that you can't get at from Perl. Let's take an example. We'll use the
1372C<$a = $b + $c> we used before, but give it a bit of context:
1373C<$b = "6XXXX"; $c = 2.3;>. Where's a good place to stop and poke around?
1374
1375What about C<pp_add>, the function we examined earlier to implement the
1376C<+> operator:
1377
1378 (gdb) break Perl_pp_add
1379 Breakpoint 1 at 0x46249f: file pp_hot.c, line 309.
1380
cea6626f 1381Notice we use C<Perl_pp_add> and not C<pp_add> - see L<perlguts/Internal Functions>.
a422fd2d
SC
1382With the breakpoint in place, we can run our program:
1383
1384 (gdb) run -e '$b = "6XXXX"; $c = 2.3; $a = $b + $c'
1385
1386Lots of junk will go past as gdb reads in the relevant source files and
1387libraries, and then:
1388
1389 Breakpoint 1, Perl_pp_add () at pp_hot.c:309
39644a26 1390 309 dSP; dATARGET; tryAMAGICbin(add,opASSIGN);
a422fd2d
SC
1391 (gdb) step
1392 311 dPOPTOPnnrl_ul;
1393 (gdb)
1394
1395We looked at this bit of code before, and we said that C<dPOPTOPnnrl_ul>
1396arranges for two C<NV>s to be placed into C<left> and C<right> - let's
1397slightly expand it:
1398
1399 #define dPOPTOPnnrl_ul NV right = POPn; \
1400 SV *leftsv = TOPs; \
1401 NV left = USE_LEFT(leftsv) ? SvNV(leftsv) : 0.0
1402
1403C<POPn> takes the SV from the top of the stack and obtains its NV either
1404directly (if C<SvNOK> is set) or by calling the C<sv_2nv> function.
1405C<TOPs> takes the next SV from the top of the stack - yes, C<POPn> uses
1406C<TOPs> - but doesn't remove it. We then use C<SvNV> to get the NV from
1407C<leftsv> in the same way as before - yes, C<POPn> uses C<SvNV>.
1408
1409Since we don't have an NV for C<$b>, we'll have to use C<sv_2nv> to
1410convert it. If we step again, we'll find ourselves there:
1411
1412 Perl_sv_2nv (sv=0xa0675d0) at sv.c:1669
1413 1669 if (!sv)
1414 (gdb)
1415
1416We can now use C<Perl_sv_dump> to investigate the SV:
1417
1418 SV = PV(0xa057cc0) at 0xa0675d0
1419 REFCNT = 1
1420 FLAGS = (POK,pPOK)
1421 PV = 0xa06a510 "6XXXX"\0
1422 CUR = 5
1423 LEN = 6
1424 $1 = void
1425
1426We know we're going to get C<6> from this, so let's finish the
1427subroutine:
1428
1429 (gdb) finish
1430 Run till exit from #0 Perl_sv_2nv (sv=0xa0675d0) at sv.c:1671
1431 0x462669 in Perl_pp_add () at pp_hot.c:311
1432 311 dPOPTOPnnrl_ul;
1433
1434We can also dump out this op: the current op is always stored in
1435C<PL_op>, and we can dump it with C<Perl_op_dump>. This'll give us
1436similar output to L<B::Debug|B::Debug>.
1437
1438 {
1439 13 TYPE = add ===> 14
1440 TARG = 1
1441 FLAGS = (SCALAR,KIDS)
1442 {
1443 TYPE = null ===> (12)
1444 (was rv2sv)
1445 FLAGS = (SCALAR,KIDS)
1446 {
1447 11 TYPE = gvsv ===> 12
1448 FLAGS = (SCALAR)
1449 GV = main::b
1450 }
1451 }
1452
10f58044 1453# finish this later #
a422fd2d
SC
1454
1455=head2 Patching
1456
1457All right, we've now had a look at how to navigate the Perl sources and
1458some things you'll need to know when fiddling with them. Let's now get
1459on and create a simple patch. Here's something Larry suggested: if a
1460C<U> is the first active format during a C<pack>, (for example,
1461C<pack "U3C8", @stuff>) then the resulting string should be treated as
1462UTF8 encoded.
1463
1464How do we prepare to fix this up? First we locate the code in question -
1465the C<pack> happens at runtime, so it's going to be in one of the F<pp>
1466files. Sure enough, C<pp_pack> is in F<pp.c>. Since we're going to be
1467altering this file, let's copy it to F<pp.c~>.
1468
a6ec74c1
JH
1469[Well, it was in F<pp.c> when this tutorial was written. It has now been
1470split off with C<pp_unpack> to its own file, F<pp_pack.c>]
1471
a422fd2d
SC
1472Now let's look over C<pp_pack>: we take a pattern into C<pat>, and then
1473loop over the pattern, taking each format character in turn into
1474C<datum_type>. Then for each possible format character, we swallow up
1475the other arguments in the pattern (a field width, an asterisk, and so
1476on) and convert the next chunk input into the specified format, adding
1477it onto the output SV C<cat>.
1478
1479How do we know if the C<U> is the first format in the C<pat>? Well, if
1480we have a pointer to the start of C<pat> then, if we see a C<U> we can
1481test whether we're still at the start of the string. So, here's where
1482C<pat> is set up:
1483
1484 STRLEN fromlen;
1485 register char *pat = SvPVx(*++MARK, fromlen);
1486 register char *patend = pat + fromlen;
1487 register I32 len;
1488 I32 datumtype;
1489 SV *fromstr;
1490
1491We'll have another string pointer in there:
1492
1493 STRLEN fromlen;
1494 register char *pat = SvPVx(*++MARK, fromlen);
1495 register char *patend = pat + fromlen;
1496 + char *patcopy;
1497 register I32 len;
1498 I32 datumtype;
1499 SV *fromstr;
1500
1501And just before we start the loop, we'll set C<patcopy> to be the start
1502of C<pat>:
1503
1504 items = SP - MARK;
1505 MARK++;
1506 sv_setpvn(cat, "", 0);
1507 + patcopy = pat;
1508 while (pat < patend) {
1509
1510Now if we see a C<U> which was at the start of the string, we turn on
1511the UTF8 flag for the output SV, C<cat>:
1512
1513 + if (datumtype == 'U' && pat==patcopy+1)
1514 + SvUTF8_on(cat);
1515 if (datumtype == '#') {
1516 while (pat < patend && *pat != '\n')
1517 pat++;
1518
1519Remember that it has to be C<patcopy+1> because the first character of
1520the string is the C<U> which has been swallowed into C<datumtype!>
1521
1522Oops, we forgot one thing: what if there are spaces at the start of the
1523pattern? C<pack(" U*", @stuff)> will have C<U> as the first active
1524character, even though it's not the first thing in the pattern. In this
1525case, we have to advance C<patcopy> along with C<pat> when we see spaces:
1526
1527 if (isSPACE(datumtype))
1528 continue;
1529
1530needs to become
1531
1532 if (isSPACE(datumtype)) {
1533 patcopy++;
1534 continue;
1535 }
1536
1537OK. That's the C part done. Now we must do two additional things before
1538this patch is ready to go: we've changed the behaviour of Perl, and so
1539we must document that change. We must also provide some more regression
1540tests to make sure our patch works and doesn't create a bug somewhere
1541else along the line.
1542
b23b8711
MS
1543The regression tests for each operator live in F<t/op/>, and so we
1544make a copy of F<t/op/pack.t> to F<t/op/pack.t~>. Now we can add our
1545tests to the end. First, we'll test that the C<U> does indeed create
1546Unicode strings.
1547
1548t/op/pack.t has a sensible ok() function, but if it didn't we could
35c336e6 1549use the one from t/test.pl.
b23b8711 1550
35c336e6
MS
1551 require './test.pl';
1552 plan( tests => 159 );
b23b8711
MS
1553
1554so instead of this:
a422fd2d
SC
1555
1556 print 'not ' unless "1.20.300.4000" eq sprintf "%vd", pack("U*",1,20,300,4000);
1557 print "ok $test\n"; $test++;
1558
35c336e6
MS
1559we can write the more sensible (see L<Test::More> for a full
1560explanation of is() and other testing functions).
b23b8711 1561
35c336e6 1562 is( "1.20.300.4000", sprintf "%vd", pack("U*",1,20,300,4000),
812f5127 1563 "U* produces unicode" );
b23b8711 1564
a422fd2d
SC
1565Now we'll test that we got that space-at-the-beginning business right:
1566
35c336e6 1567 is( "1.20.300.4000", sprintf "%vd", pack(" U*",1,20,300,4000),
812f5127 1568 " with spaces at the beginning" );
a422fd2d
SC
1569
1570And finally we'll test that we don't make Unicode strings if C<U> is B<not>
1571the first active format:
1572
35c336e6 1573 isnt( v1.20.300.4000, sprintf "%vd", pack("C0U*",1,20,300,4000),
812f5127 1574 "U* not first isn't unicode" );
a422fd2d 1575
35c336e6
MS
1576Mustn't forget to change the number of tests which appears at the top,
1577or else the automated tester will get confused. This will either look
1578like this:
a422fd2d 1579
35c336e6
MS
1580 print "1..156\n";
1581
1582or this:
1583
1584 plan( tests => 156 );
a422fd2d
SC
1585
1586We now compile up Perl, and run it through the test suite. Our new
1587tests pass, hooray!
1588
1589Finally, the documentation. The job is never done until the paperwork is
1590over, so let's describe the change we've just made. The relevant place
1591is F<pod/perlfunc.pod>; again, we make a copy, and then we'll insert
1592this text in the description of C<pack>:
1593
1594 =item *
1595
1596 If the pattern begins with a C<U>, the resulting string will be treated
1597 as Unicode-encoded. You can force UTF8 encoding on in a string with an
1598 initial C<U0>, and the bytes that follow will be interpreted as Unicode
1599 characters. If you don't want this to happen, you can begin your pattern
1600 with C<C0> (or anything else) to force Perl not to UTF8 encode your
1601 string, and then follow this with a C<U*> somewhere in your pattern.
1602
1603All done. Now let's create the patch. F<Porting/patching.pod> tells us
1604that if we're making major changes, we should copy the entire directory
1605to somewhere safe before we begin fiddling, and then do
13a2d996 1606
a422fd2d
SC
1607 diff -ruN old new > patch
1608
1609However, we know which files we've changed, and we can simply do this:
1610
1611 diff -u pp.c~ pp.c > patch
1612 diff -u t/op/pack.t~ t/op/pack.t >> patch
1613 diff -u pod/perlfunc.pod~ pod/perlfunc.pod >> patch
1614
1615We end up with a patch looking a little like this:
1616
1617 --- pp.c~ Fri Jun 02 04:34:10 2000
1618 +++ pp.c Fri Jun 16 11:37:25 2000
1619 @@ -4375,6 +4375,7 @@
1620 register I32 items;
1621 STRLEN fromlen;
1622 register char *pat = SvPVx(*++MARK, fromlen);
1623 + char *patcopy;
1624 register char *patend = pat + fromlen;
1625 register I32 len;
1626 I32 datumtype;
1627 @@ -4405,6 +4406,7 @@
1628 ...
1629
1630And finally, we submit it, with our rationale, to perl5-porters. Job
1631done!
1632
f7e1e956
MS
1633=head2 Patching a core module
1634
1635This works just like patching anything else, with an extra
1636consideration. Many core modules also live on CPAN. If this is so,
1637patch the CPAN version instead of the core and send the patch off to
1638the module maintainer (with a copy to p5p). This will help the module
1639maintainer keep the CPAN version in sync with the core version without
1640constantly scanning p5p.
1641
acbe17fc
JP
1642=head2 Adding a new function to the core
1643
1644If, as part of a patch to fix a bug, or just because you have an
1645especially good idea, you decide to add a new function to the core,
1646discuss your ideas on p5p well before you start work. It may be that
1647someone else has already attempted to do what you are considering and
1648can give lots of good advice or even provide you with bits of code
1649that they already started (but never finished).
1650
1651You have to follow all of the advice given above for patching. It is
1652extremely important to test any addition thoroughly and add new tests
1653to explore all boundary conditions that your new function is expected
1654to handle. If your new function is used only by one module (e.g. toke),
1655then it should probably be named S_your_function (for static); on the
210b36aa 1656other hand, if you expect it to accessible from other functions in
acbe17fc
JP
1657Perl, you should name it Perl_your_function. See L<perlguts/Internal Functions>
1658for more details.
1659
1660The location of any new code is also an important consideration. Don't
1661just create a new top level .c file and put your code there; you would
1662have to make changes to Configure (so the Makefile is created properly),
1663as well as possibly lots of include files. This is strictly pumpking
1664business.
1665
1666It is better to add your function to one of the existing top level
1667source code files, but your choice is complicated by the nature of
1668the Perl distribution. Only the files that are marked as compiled
1669static are located in the perl executable. Everything else is located
1670in the shared library (or DLL if you are running under WIN32). So,
1671for example, if a function was only used by functions located in
1672toke.c, then your code can go in toke.c. If, however, you want to call
1673the function from universal.c, then you should put your code in another
1674location, for example util.c.
1675
1676In addition to writing your c-code, you will need to create an
1677appropriate entry in embed.pl describing your function, then run
1678'make regen_headers' to create the entries in the numerous header
1679files that perl needs to compile correctly. See L<perlguts/Internal Functions>
1680for information on the various options that you can set in embed.pl.
1681You will forget to do this a few (or many) times and you will get
1682warnings during the compilation phase. Make sure that you mention
1683this when you post your patch to P5P; the pumpking needs to know this.
1684
1685When you write your new code, please be conscious of existing code
884bad00 1686conventions used in the perl source files. See L<perlstyle> for
acbe17fc
JP
1687details. Although most of the guidelines discussed seem to focus on
1688Perl code, rather than c, they all apply (except when they don't ;).
1689See also I<Porting/patching.pod> file in the Perl source distribution
1690for lots of details about both formatting and submitting patches of
1691your changes.
1692
1693Lastly, TEST TEST TEST TEST TEST any code before posting to p5p.
1694Test on as many platforms as you can find. Test as many perl
1695Configure options as you can (e.g. MULTIPLICITY). If you have
1696profiling or memory tools, see L<EXTERNAL TOOLS FOR DEBUGGING PERL>
210b36aa 1697below for how to use them to further test your code. Remember that
acbe17fc
JP
1698most of the people on P5P are doing this on their own time and
1699don't have the time to debug your code.
f7e1e956
MS
1700
1701=head2 Writing a test
1702
1703Every module and built-in function has an associated test file (or
1704should...). If you add or change functionality, you have to write a
1705test. If you fix a bug, you have to write a test so that bug never
1706comes back. If you alter the docs, it would be nice to test what the
1707new documentation says.
1708
1709In short, if you submit a patch you probably also have to patch the
1710tests.
1711
1712For modules, the test file is right next to the module itself.
1713F<lib/strict.t> tests F<lib/strict.pm>. This is a recent innovation,
1714so there are some snags (and it would be wonderful for you to brush
1715them out), but it basically works that way. Everything else lives in
1716F<t/>.
1717
1718=over 3
1719
1720=item F<t/base/>
1721
1722Testing of the absolute basic functionality of Perl. Things like
1723C<if>, basic file reads and writes, simple regexes, etc. These are
1724run first in the test suite and if any of them fail, something is
1725I<really> broken.
1726
1727=item F<t/cmd/>
1728
1729These test the basic control structures, C<if/else>, C<while>,
35c336e6 1730subroutines, etc.
f7e1e956
MS
1731
1732=item F<t/comp/>
1733
1734Tests basic issues of how Perl parses and compiles itself.
1735
1736=item F<t/io/>
1737
1738Tests for built-in IO functions, including command line arguments.
1739
1740=item F<t/lib/>
1741
1742The old home for the module tests, you shouldn't put anything new in
1743here. There are still some bits and pieces hanging around in here
1744that need to be moved. Perhaps you could move them? Thanks!
1745
1746=item F<t/op/>
1747
1748Tests for perl's built in functions that don't fit into any of the
1749other directories.
1750
1751=item F<t/pod/>
1752
1753Tests for POD directives. There are still some tests for the Pod
1754modules hanging around in here that need to be moved out into F<lib/>.
1755
1756=item F<t/run/>
1757
1758Testing features of how perl actually runs, including exit codes and
1759handling of PERL* environment variables.
1760
1761=back
1762
1763The core uses the same testing style as the rest of Perl, a simple
1764"ok/not ok" run through Test::Harness, but there are a few special
1765considerations.
1766
35c336e6
MS
1767There are three ways to write a test in the core. Test::More,
1768t/test.pl and ad hoc C<print $test ? "ok 42\n" : "not ok 42\n">. The
1769decision of which to use depends on what part of the test suite you're
1770working on. This is a measure to prevent a high-level failure (such
1771as Config.pm breaking) from causing basic functionality tests to fail.
1772
1773=over 4
1774
1775=item t/base t/comp
1776
1777Since we don't know if require works, or even subroutines, use ad hoc
1778tests for these two. Step carefully to avoid using the feature being
1779tested.
1780
1781=item t/cmd t/run t/io t/op
1782
1783Now that basic require() and subroutines are tested, you can use the
1784t/test.pl library which emulates the important features of Test::More
1785while using a minimum of core features.
1786
1787You can also conditionally use certain libraries like Config, but be
1788sure to skip the test gracefully if it's not there.
1789
1790=item t/lib ext lib
1791
1792Now that the core of Perl is tested, Test::More can be used. You can
1793also use the full suite of core modules in the tests.
1794
1795=back
f7e1e956
MS
1796
1797When you say "make test" Perl uses the F<t/TEST> program to run the
1798test suite. All tests are run from the F<t/> directory, B<not> the
1799directory which contains the test. This causes some problems with the
1800tests in F<lib/>, so here's some opportunity for some patching.
1801
1802You must be triply conscious of cross-platform concerns. This usually
1803boils down to using File::Spec and avoiding things like C<fork()> and
1804C<system()> unless absolutely necessary.
1805
1806
902b9dbf
MF
1807=head1 EXTERNAL TOOLS FOR DEBUGGING PERL
1808
1809Sometimes it helps to use external tools while debugging and
1810testing Perl. This section tries to guide you through using
1811some common testing and debugging tools with Perl. This is
1812meant as a guide to interfacing these tools with Perl, not
1813as any kind of guide to the use of the tools themselves.
1814
1815=head2 Rational Software's Purify
1816
1817Purify is a commercial tool that is helpful in identifying
1818memory overruns, wild pointers, memory leaks and other such
1819badness. Perl must be compiled in a specific way for
1820optimal testing with Purify. Purify is available under
1821Windows NT, Solaris, HP-UX, SGI, and Siemens Unix.
1822
1823The only currently known leaks happen when there are
1824compile-time errors within eval or require. (Fixing these
1825is non-trivial, unfortunately, but they must be fixed
1826eventually.)
1827
1828=head2 Purify on Unix
1829
1830On Unix, Purify creates a new Perl binary. To get the most
1831benefit out of Purify, you should create the perl to Purify
1832using:
1833
1834 sh Configure -Accflags=-DPURIFY -Doptimize='-g' \
1835 -Uusemymalloc -Dusemultiplicity
1836
1837where these arguments mean:
1838
1839=over 4
1840
1841=item -Accflags=-DPURIFY
1842
1843Disables Perl's arena memory allocation functions, as well as
1844forcing use of memory allocation functions derived from the
1845system malloc.
1846
1847=item -Doptimize='-g'
1848
1849Adds debugging information so that you see the exact source
1850statements where the problem occurs. Without this flag, all
1851you will see is the source filename of where the error occurred.
1852
1853=item -Uusemymalloc
1854
1855Disable Perl's malloc so that Purify can more closely monitor
1856allocations and leaks. Using Perl's malloc will make Purify
1857report most leaks in the "potential" leaks category.
1858
1859=item -Dusemultiplicity
1860
1861Enabling the multiplicity option allows perl to clean up
1862thoroughly when the interpreter shuts down, which reduces the
1863number of bogus leak reports from Purify.
1864
1865=back
1866
1867Once you've compiled a perl suitable for Purify'ing, then you
1868can just:
1869
1870 make pureperl
1871
1872which creates a binary named 'pureperl' that has been Purify'ed.
1873This binary is used in place of the standard 'perl' binary
1874when you want to debug Perl memory problems.
1875
1f56d61a
JH
1876To minimize the number of memory leak false alarms
1877(see L</PERL_DESTRUCT_LEVEL>), set environment variable
1878PERL_DESTRUCT_LEVEL to 2.
1879
1880 setenv PERL_DESTRUCT_LEVEL 2
1881
1882In Bourne-type shells:
1883
1884 PERL_DESTRUCT_LEVEL=2
1885 export PERL_DESTRUCT_LEVEL
1886
902b9dbf
MF
1887As an example, to show any memory leaks produced during the
1888standard Perl testset you would create and run the Purify'ed
1889perl as:
1890
1891 make pureperl
1892 cd t
1893 ../pureperl -I../lib harness
1894
1895which would run Perl on test.pl and report any memory problems.
1896
1897Purify outputs messages in "Viewer" windows by default. If
1898you don't have a windowing environment or if you simply
1899want the Purify output to unobtrusively go to a log file
1900instead of to the interactive window, use these following
1901options to output to the log file "perl.log":
1902
1903 setenv PURIFYOPTIONS "-chain-length=25 -windows=no \
1904 -log-file=perl.log -append-logfile=yes"
1905
1906If you plan to use the "Viewer" windows, then you only need this option:
1907
1908 setenv PURIFYOPTIONS "-chain-length=25"
1909
c406981e
JH
1910In Bourne-type shells:
1911
1912 PURIFY_OPTIONS="..."
1913 export PURIFY_OPTIONS
1914
1915or if you have the "env" utility:
1916
1917 env PURIFY_OPTIONS="..." ../pureperl ...
1918
902b9dbf
MF
1919=head2 Purify on NT
1920
1921Purify on Windows NT instruments the Perl binary 'perl.exe'
1922on the fly. There are several options in the makefile you
1923should change to get the most use out of Purify:
1924
1925=over 4
1926
1927=item DEFINES
1928
1929You should add -DPURIFY to the DEFINES line so the DEFINES
1930line looks something like:
1931
1932 DEFINES = -DWIN32 -D_CONSOLE -DNO_STRICT $(CRYPT_FLAG) -DPURIFY=1
1933
1934to disable Perl's arena memory allocation functions, as
1935well as to force use of memory allocation functions derived
1936from the system malloc.
1937
1938=item USE_MULTI = define
1939
1940Enabling the multiplicity option allows perl to clean up
1941thoroughly when the interpreter shuts down, which reduces the
1942number of bogus leak reports from Purify.
1943
1944=item #PERL_MALLOC = define
1945
1946Disable Perl's malloc so that Purify can more closely monitor
1947allocations and leaks. Using Perl's malloc will make Purify
1948report most leaks in the "potential" leaks category.
1949
1950=item CFG = Debug
1951
1952Adds debugging information so that you see the exact source
1953statements where the problem occurs. Without this flag, all
1954you will see is the source filename of where the error occurred.
1955
1956=back
1957
1958As an example, to show any memory leaks produced during the
1959standard Perl testset you would create and run Purify as:
1960
1961 cd win32
1962 make
1963 cd ../t
1964 purify ../perl -I../lib harness
1965
1966which would instrument Perl in memory, run Perl on test.pl,
1967then finally report any memory problems.
1968
09187cb1
JH
1969=head2 Compaq's/Digital's Third Degree
1970
1971Third Degree is a tool for memory leak detection and memory access checks.
1972It is one of the many tools in the ATOM toolkit. The toolkit is only
1973available on Tru64 (formerly known as Digital UNIX formerly known as
1974DEC OSF/1).
1975
1976When building Perl, you must first run Configure with -Doptimize=-g
1977and -Uusemymalloc flags, after that you can use the make targets
51a35ef1
JH
1978"perl.third" and "test.third". (What is required is that Perl must be
1979compiled using the C<-g> flag, you may need to re-Configure.)
09187cb1 1980
64cea5fd 1981The short story is that with "atom" you can instrument the Perl
83f0ef60 1982executable to create a new executable called F<perl.third>. When the
4ae3d70a 1983instrumented executable is run, it creates a log of dubious memory
83f0ef60 1984traffic in file called F<perl.3log>. See the manual pages of atom and
4ae3d70a
JH
1985third for more information. The most extensive Third Degree
1986documentation is available in the Compaq "Tru64 UNIX Programmer's
1987Guide", chapter "Debugging Programs with Third Degree".
64cea5fd 1988
83f0ef60 1989The "test.third" leaves a lot of files named F<perl.3log.*> in the t/
64cea5fd
JH
1990subdirectory. There is a problem with these files: Third Degree is so
1991effective that it finds problems also in the system libraries.
83f0ef60
JH
1992Therefore there are certain types of errors that you should ignore in
1993your debugging. Errors with stack traces matching
64cea5fd
JH
1994
1995 __actual_atof|__catgets|_doprnt|__exc_|__exec|_findio|__localtime|setlocale|__sia_|__strxfrm
1996
1997(all in libc.so) are known to be non-serious. You can also
1998ignore the combinations
1999
2000 Perl_gv_fetchfile() calling strcpy()
2001 S_doopen_pmc() calling strcmp()
2002
2003causing "rih" (reading invalid heap) errors.
2004
2005There are also leaks that for given certain definition of a leak,
2006aren't. See L</PERL_DESTRUCT_LEVEL> for more information.
2007
2008=head2 PERL_DESTRUCT_LEVEL
2009
2010If you want to run any of the tests yourself manually using the
2011pureperl or perl.third executables, please note that by default
2012perl B<does not> explicitly cleanup all the memory it has allocated
2013(such as global memory arenas) but instead lets the exit() of
2014the whole program "take care" of such allocations, also known
2015as "global destruction of objects".
2016
2017There is a way to tell perl to do complete cleanup: set the
2018environment variable PERL_DESTRUCT_LEVEL to a non-zero value.
2019The t/TEST wrapper does set this to 2, and this is what you
2020need to do too, if you don't want to see the "global leaks":
1f56d61a 2021For example, for "third-degreed" Perl:
64cea5fd 2022
1f56d61a 2023 env PERL_DESTRUCT_LEVEL=2 ./perl.third -Ilib t/foo/bar.t
09187cb1 2024
414f2397
RGS
2025(Note: the mod_perl apache module uses also this environment variable
2026for its own purposes and extended its semantics. Refer to the mod_perl
2027documentation for more information.)
2028
51a35ef1
JH
2029=head2 Profiling
2030
2031Depending on your platform there are various of profiling Perl.
2032
2033There are two commonly used techniques of profiling executables:
10f58044 2034I<statistical time-sampling> and I<basic-block counting>.
51a35ef1
JH
2035
2036The first method takes periodically samples of the CPU program
2037counter, and since the program counter can be correlated with the code
2038generated for functions, we get a statistical view of in which
2039functions the program is spending its time. The caveats are that very
2040small/fast functions have lower probability of showing up in the
2041profile, and that periodically interrupting the program (this is
2042usually done rather frequently, in the scale of milliseconds) imposes
2043an additional overhead that may skew the results. The first problem
2044can be alleviated by running the code for longer (in general this is a
2045good idea for profiling), the second problem is usually kept in guard
2046by the profiling tools themselves.
2047
10f58044 2048The second method divides up the generated code into I<basic blocks>.
51a35ef1
JH
2049Basic blocks are sections of code that are entered only in the
2050beginning and exited only at the end. For example, a conditional jump
2051starts a basic block. Basic block profiling usually works by
10f58044 2052I<instrumenting> the code by adding I<enter basic block #nnnn>
51a35ef1
JH
2053book-keeping code to the generated code. During the execution of the
2054code the basic block counters are then updated appropriately. The
2055caveat is that the added extra code can skew the results: again, the
2056profiling tools usually try to factor their own effects out of the
2057results.
2058
83f0ef60
JH
2059=head2 Gprof Profiling
2060
51a35ef1
JH
2061gprof is a profiling tool available in many UNIX platforms,
2062it uses F<statistical time-sampling>.
83f0ef60
JH
2063
2064You can build a profiled version of perl called "perl.gprof" by
51a35ef1
JH
2065invoking the make target "perl.gprof" (What is required is that Perl
2066must be compiled using the C<-pg> flag, you may need to re-Configure).
2067Running the profiled version of Perl will create an output file called
2068F<gmon.out> is created which contains the profiling data collected
2069during the execution.
83f0ef60
JH
2070
2071The gprof tool can then display the collected data in various ways.
2072Usually gprof understands the following options:
2073
2074=over 4
2075
2076=item -a
2077
2078Suppress statically defined functions from the profile.
2079
2080=item -b
2081
2082Suppress the verbose descriptions in the profile.
2083
2084=item -e routine
2085
2086Exclude the given routine and its descendants from the profile.
2087
2088=item -f routine
2089
2090Display only the given routine and its descendants in the profile.
2091
2092=item -s
2093
2094Generate a summary file called F<gmon.sum> which then may be given
2095to subsequent gprof runs to accumulate data over several runs.
2096
2097=item -z
2098
2099Display routines that have zero usage.
2100
2101=back
2102
2103For more detailed explanation of the available commands and output
2104formats, see your own local documentation of gprof.
2105
51a35ef1
JH
2106=head2 GCC gcov Profiling
2107
10f58044 2108Starting from GCC 3.0 I<basic block profiling> is officially available
51a35ef1
JH
2109for the GNU CC.
2110
2111You can build a profiled version of perl called F<perl.gcov> by
2112invoking the make target "perl.gcov" (what is required that Perl must
2113be compiled using gcc with the flags C<-fprofile-arcs
2114-ftest-coverage>, you may need to re-Configure).
2115
2116Running the profiled version of Perl will cause profile output to be
2117generated. For each source file an accompanying ".da" file will be
2118created.
2119
2120To display the results you use the "gcov" utility (which should
2121be installed if you have gcc 3.0 or newer installed). F<gcov> is
2122run on source code files, like this
2123
2124 gcov sv.c
2125
2126which will cause F<sv.c.gcov> to be created. The F<.gcov> files
2127contain the source code annotated with relative frequencies of
2128execution indicated by "#" markers.
2129
2130Useful options of F<gcov> include C<-b> which will summarise the
2131basic block, branch, and function call coverage, and C<-c> which
2132instead of relative frequencies will use the actual counts. For
2133more information on the use of F<gcov> and basic block profiling
2134with gcc, see the latest GNU CC manual, as of GCC 3.0 see
2135
2136 http://gcc.gnu.org/onlinedocs/gcc-3.0/gcc.html
2137
2138and its section titled "8. gcov: a Test Coverage Program"
2139
2140 http://gcc.gnu.org/onlinedocs/gcc-3.0/gcc_8.html#SEC132
2141
4ae3d70a
JH
2142=head2 Pixie Profiling
2143
51a35ef1
JH
2144Pixie is a profiling tool available on IRIX and Tru64 (aka Digital
2145UNIX aka DEC OSF/1) platforms. Pixie does its profiling using
10f58044 2146I<basic-block counting>.
4ae3d70a 2147
83f0ef60 2148You can build a profiled version of perl called F<perl.pixie> by
51a35ef1
JH
2149invoking the make target "perl.pixie" (what is required is that Perl
2150must be compiled using the C<-g> flag, you may need to re-Configure).
2151
2152In Tru64 a file called F<perl.Addrs> will also be silently created,
2153this file contains the addresses of the basic blocks. Running the
2154profiled version of Perl will create a new file called "perl.Counts"
2155which contains the counts for the basic block for that particular
2156program execution.
4ae3d70a 2157
51a35ef1 2158To display the results you use the F<prof> utility. The exact
4ae3d70a
JH
2159incantation depends on your operating system, "prof perl.Counts" in
2160IRIX, and "prof -pixie -all -L. perl" in Tru64.
2161
6c41479b
JH
2162In IRIX the following prof options are available:
2163
2164=over 4
2165
2166=item -h
2167
2168Reports the most heavily used lines in descending order of use.
6e36760b 2169Useful for finding the hotspot lines.
6c41479b
JH
2170
2171=item -l
2172
2173Groups lines by procedure, with procedures sorted in descending order of use.
2174Within a procedure, lines are listed in source order.
6e36760b 2175Useful for finding the hotspots of procedures.
6c41479b
JH
2176
2177=back
2178
2179In Tru64 the following options are available:
2180
2181=over 4
2182
3958b146 2183=item -p[rocedures]
6c41479b 2184
3958b146 2185Procedures sorted in descending order by the number of cycles executed
6e36760b 2186in each procedure. Useful for finding the hotspot procedures.
6c41479b
JH
2187(This is the default option.)
2188
24000d2f 2189=item -h[eavy]
6c41479b 2190
6e36760b
JH
2191Lines sorted in descending order by the number of cycles executed in
2192each line. Useful for finding the hotspot lines.
6c41479b 2193
24000d2f 2194=item -i[nvocations]
6c41479b 2195
6e36760b
JH
2196The called procedures are sorted in descending order by number of calls
2197made to the procedures. Useful for finding the most used procedures.
6c41479b 2198
24000d2f 2199=item -l[ines]
6c41479b
JH
2200
2201Grouped by procedure, sorted by cycles executed per procedure.
6e36760b 2202Useful for finding the hotspots of procedures.
6c41479b
JH
2203
2204=item -testcoverage
2205
2206The compiler emitted code for these lines, but the code was unexecuted.
2207
24000d2f 2208=item -z[ero]
6c41479b
JH
2209
2210Unexecuted procedures.
2211
aa500c9e 2212=back
6c41479b
JH
2213
2214For further information, see your system's manual pages for pixie and prof.
4ae3d70a 2215
b8ddf6b3
SB
2216=head2 Miscellaneous tricks
2217
2218=over 4
2219
2220=item *
2221
cc177e1a 2222Those debugging perl with the DDD frontend over gdb may find the
b8ddf6b3
SB
2223following useful:
2224
2225You can extend the data conversion shortcuts menu, so for example you
2226can display an SV's IV value with one click, without doing any typing.
2227To do that simply edit ~/.ddd/init file and add after:
2228
2229 ! Display shortcuts.
2230 Ddd*gdbDisplayShortcuts: \
2231 /t () // Convert to Bin\n\
2232 /d () // Convert to Dec\n\
2233 /x () // Convert to Hex\n\
2234 /o () // Convert to Oct(\n\
2235
2236the following two lines:
2237
2238 ((XPV*) (())->sv_any )->xpv_pv // 2pvx\n\
2239 ((XPVIV*) (())->sv_any )->xiv_iv // 2ivx
2240
2241so now you can do ivx and pvx lookups or you can plug there the
2242sv_peek "conversion":
2243
2244 Perl_sv_peek(my_perl, (SV*)()) // sv_peek
2245
2246(The my_perl is for threaded builds.)
2247Just remember that every line, but the last one, should end with \n\
2248
2249Alternatively edit the init file interactively via:
22503rd mouse button -> New Display -> Edit Menu
2251
2252Note: you can define up to 20 conversion shortcuts in the gdb
2253section.
2254
2255=back
2256
a422fd2d
SC
2257=head2 CONCLUSION
2258
2259We've had a brief look around the Perl source, an overview of the stages
2260F<perl> goes through when it's running your code, and how to use a
902b9dbf
MF
2261debugger to poke at the Perl guts. We took a very simple problem and
2262demonstrated how to solve it fully - with documentation, regression
2263tests, and finally a patch for submission to p5p. Finally, we talked
2264about how to use external tools to debug and test Perl.
a422fd2d
SC
2265
2266I'd now suggest you read over those references again, and then, as soon
2267as possible, get your hands dirty. The best way to learn is by doing,
2268so:
2269
2270=over 3
2271
2272=item *
2273
2274Subscribe to perl5-porters, follow the patches and try and understand
2275them; don't be afraid to ask if there's a portion you're not clear on -
2276who knows, you may unearth a bug in the patch...
2277
2278=item *
2279
2280Keep up to date with the bleeding edge Perl distributions and get
2281familiar with the changes. Try and get an idea of what areas people are
2282working on and the changes they're making.
2283
2284=item *
2285
3e148164 2286Do read the README associated with your operating system, e.g. README.aix
a1f349fd
MB
2287on the IBM AIX OS. Don't hesitate to supply patches to that README if
2288you find anything missing or changed over a new OS release.
2289
2290=item *
2291
a422fd2d
SC
2292Find an area of Perl that seems interesting to you, and see if you can
2293work out how it works. Scan through the source, and step over it in the
2294debugger. Play, poke, investigate, fiddle! You'll probably get to
2295understand not just your chosen area but a much wider range of F<perl>'s
2296activity as well, and probably sooner than you'd think.
2297
2298=back
2299
2300=over 3
2301
2302=item I<The Road goes ever on and on, down from the door where it began.>
2303
2304=back
2305
2306If you can do these things, you've started on the long road to Perl porting.
2307Thanks for wanting to help make Perl better - and happy hacking!
2308
e8cd7eae
GS
2309=head1 AUTHOR
2310
2311This document was written by Nathan Torkington, and is maintained by
2312the perl5-porters mailing list.
2313